In order to control this potential confounding factor, we analysed HTC the subgroup of 87 patients adequately treated within 24 hours after VAP diagnosis. In the multivariate analysis, the only risk factor of ICU or hospital death was having at least one chronic illness, not piperacillin resistance.Other studies comparing outcomes of susceptible and resistant P. aeruginosa infections are scarce. Recently, Combes et al. conducted a post-hoc analysis of the PNEUMA randomized study, which involved only cases of adequately treated PA VAP [9]. PRPA-VAP was associated with a higher mortality than PSPA-VAP. However, after adjustment for age, female gender, underlying comorbidities, and SOFA score, piperacillin resistance was no longer associated with mortality (OR, 2.00; 95% CI, 0.72 to 5.61; P = 0.

194). Similarly, in a retrospective cohort study evaluating the epidemiological characteristics of 135 episodes of VAP caused by PRPA or PSPA, Trouillet et al. did not find any increased death rates for PRPA infections [10]. Data from the few studies concerning P.aeruginosa bacteraemia reported similar results. Blot et al. conducted a retrospective study on antibiotic-susceptible and antibiotic-resistant nosocomial gram-negative bacilli bacteraemia in critically ill patients. Factors associated with in-hospital mortality were the age, a high-risk source of bacteraemia, and APACHE II score, but not antibiotic resistance [28]. This result is in accordance with other studies conducted about gram-negative bacilli infection [29], as well as Staphylococcus aureus pneumonia [30].

Some studies have described higher mortality rates in association with infections caused by antibiotic resistant P. aeruginosa. In a retrospective matched control study by Aloush et al., patients infected with multi-resistant P. aeruginosa strains had higher mortality rates and increased durations of hospital stay, but the control patients did not have P. aeruginosa infections [31]. In an older study evaluating health and economic outcomes of resistant Pseudomonas infections, the same group found that, although the patients with resistant strain at hospital admission did not have a poorer overall prognosis, the emergence of resistant strains during the hospital stay was associated with a prolonged length of stay and a higher hospital mortality rate [32].Our study highlighted the major role of previous antibiotic therapy.

The emergence of PRPA: PRPA VAP episodes were Batimastat significantly associated with the administration of broad-spectrum antimicrobials at admission to ICU, such as ureidopenicillins, carboxypenicillins or fluoroquinolones. These findings are in accordance with the results of Harris et al. [33], who found that the piperacillin-tazobactam exposure was the major factor that predispose for the development of an infection with a multidrug resistant P. aeruginosa. In a clinical study, Reinhardt et al. found that resistant P.